Civilian casualties can be prevented with better use of drones. By Sarah Holewinski and Larry Lewis
The way proponents talk about them, you’d think Zeus gifted drones straight from his backpack. Alternatively, you might also believe that drones herald the age of robotic, soulless warfare and are evil in their very nature.
We’re watching the renewed drone debate this week, much of which is focused on whether drones are legal and justified -- a critical discussion to have, but there’s another one, which is how the use of drones can be made more responsible in the immediate term.
Drones aren’t magical. Any weapon system is only as effective as how it’s used. In that spirit, we offer here some practical steps Washington could immediately take to help drones meet President Obama’s criteria from his May speech, when he said, “Before any strike is taken, there must be near-certainty that no civilians will be killed or injured -- the highest standard we can set.”
We believe the United States can live up to this admirable standard with deliberate effort. The good news is that we know it’s possible because it’s been done before. After taking a public bashing over civilian casualties in Afghanistan, U.S. commanders made a point of investigating, analyzing, responding to and learning lessons from civilian harm. They looked at the problem of civilian harm square in the eye and addressed it. We ask the Obama administration to do the same when using drones.
Don’t equate precision with low civilian casualties
“Precision” is the new black. The administration touts drones as the most precise weapon in history. But a drone isn’t precise in and of itself, and using this word incorrectly diverts needed discussions about gaps in policy or operations that risk civilian harm. Consider that a Center for Naval Analyses study in Afghanistan found that drone strikes were ten times more likely to cause casualties than strikes by manned aircraft. Why? Because of gaps in training, communications, and intelligence.
Our point is that it’s not about the drone itself. Drones have weaknesses, just like any other weapon. For a drone to be more precise than a fighter jet, it needs the right inputs: quality intelligence, effective communications among the many players and knowledge of pattern of life. A big part of minimizing civilian harm is the operator knowing the identity of the target and whoever else is in the area. Video feeds can’t see into buildings or cars, and while “pattern of life” surveillance from the sky can give additional insight, this is by no means perfect.
Drones may reduce civilian risk compared to massive air campaigns or invading ground forces, but these aren’t valid alternatives to fight terrorism. Alternatives to drones appear to be on the docket with recent special operations raids in Libya and Somalia, and this is a good thing. Special operations forces in Somalia stopped their raid because they saw women inside the building. A drone might have killed everyone inside. And of course, the State Department should have a big stake in other counterterrorism approaches, if only because drone strikes can get in the way of diplomacy with needed security partners.
Know the real reality, not the desired one
To avoid killing civilians to “near-certainty” means the U.S. needs to know how many civilians may be harmed. The reality is that there are big challenges to getting that information both before and after an air strike. Video feeds, often the primary intelligence option in a place like Pakistan, don’t show a civilian hiding in the basement of a home or an al Qaeda operative’s wife. While it has been said that independent assessments of civilian casualties tend to be too high, official U.S. estimates are certainly too low. The truth is likely to be somewhere in between.
Civilian casualty data analysis can show root causes of each accident and point to ways that operations can be less harmful in the future. In Afghanistan, this learning process was more important for reducing civilian harm than any individual weapon system. Yet this analysis and learning doesn’t happen in Pakistan, Yemen or Somalia.
It’s time the U.S. relearn what it did in Afghanistan and formally review the civilian impact of drone strikes, so that the program is informed by facts not hopes. A review like this, which is performed constantly on other issues, is likely to highlight operational practices that put civilians at more risk than necessary.
Take the CIA out of the lead on drones
The CIA will always have a supporting role in drone operations, but the military appears better suited for the command role. The drone program would benefit from long-established doctrine and recently improved processes for targeting and reducing civilian harm. And the military has a history of openly debating the ethical use of force.
To his credit, President Obama pledged to transfer drone command from the CIA to the Department of Defense. Staffers on the Hill tell us that the process is slow going due to worries about DOD’s capacity to handle the entire drone program, and worry that special operations forces often act as a mini-CIA within the military without proper oversight. These are valid concerns, but they can be addressed. The Armed Services Committees should be proactive in solving the capacity and accountability issues. For example, military commanders leading a drone program could give a progress report to these committees every six months.
Get thee a “civilian” mindset
Any Air Force pilot who recently served in Afghanistan will tell you that the “civilian” is the center of gravity. Civilian casualties were simply not acceptable. Starting with Gen. Stanley McChrystal, former International Security Assistance Force chief, commanders drilled this mindset into their troops, understanding that civilian harm and anger crippled the mission. Better training and doctrine on avoiding civilians and responding properly to harm followed for all deploying forces. In our visits to bases both stateside and in Afghanistan, we saw a sea change for the better in how military personnel thought about civilian harm. The isolation of drone pilots from this shift in mindset means that new pilots do not benefit from an updated version of McChrystal’s shakedown on killing civilians And, the rapid expansion of the drone program means training time for these issues is scarce.
Don’t walk away from civilian harm that happens
From Roman to modern combat operations, civilian harm is an unfailing fact. Drone technology will not change this. Even with all the right inputs and training, there’s what one pilot called the "oops” factor. Whether casualties are in the single digits or hundreds, Washington should have some plan to address the inevitable.
The U.S. has offered no channels for civilians harmed by drone strikes outside of declared theaters of conflict to come forward with their grievances or receive help. What’s striking is that the U.S. military in Vietnam, Korea, Afghanistan and Iraq did recognize civilian harm with great effect. Locals tended to view U.S. operations more favorably for making the effort. Not having troops on the ground where many drones strike doesn’t mean the U.S. should just walk away from civilian harm; doing so can cause long lasting resentment and invite criticism about American values. Partner states can investigate incidents of civilian harm. The State Department can broker assistance agreements with local officials. USAID programs can offer longer-term assistance to victims. Aid can even be channeled through the host government in some cases. Our point is that there are practical solutions to be found, if policymakers prioritize the principle.
Sarah Holewinski is executive director of the Center for Civilians in Conflict. Larry Lewis is a research analyst and field representative for CNA’s Operations Evaluation Group. The opinions expressed are the authors' alone.